A Sensor-Driven Optimization Framework for Asset Management in Energy Systems: Implications for Full and Partial Digital Transformation in Hydro Fleets

This paper proposes a novel prognostics-driven approach to optimize operations and maintenance (O&M) decisions in hydropower systems. Our approach harnesses the insights from sensor data to accurately predict the remaining lifetime distribution of critical generation assets in hydropower systems, i.e., thrust bearings, and use these predictions to optimally schedule O&M actions for a fleet of hydro generators. We consider complex interdependencies across hydro generator failure risks, reservoir, production, and demand management decisions. We propose a stochastic joint O&M scheduling model to tackle the unique challenges of hydropower O&M including the interdependency of generation capacities, the nonlinear nature of power production, operational requirements, and uncertainties. We develop a two-level decomposition-based solution algorithm to effectively handle large-scale cases. The algorithm incorporates a combination of Benders optimality cuts and integer cuts to solve the problem in an efficient manner. We design an experimental framework to evaluate the proposed prognostics-driven O&M scheduling framework, using real-world condition monitoring data from hydropower systems, historical market prices, and water inflow data. The developed framework can be partially implemented for a phased-in approach. Our experiments demonstrate the significant benefits of the sensor-driven O&M framework in improving reliability, availability, effective usage of resources, and system profitability, especially when gradually shifting from traditional time-based maintenance policies to condition-based prognostics-driven maintenance policies.